Over 20 years ago, William Riker proposed that the study of electoral systems and their consequences constituted the best-developed research program in contemporary political science. Scores of scholars followed Maurice Duverger's lead, emphasizing how institutions affect the strategic environment facing political actors and the contours of party systems. Other researchers followed Seymour Martin Lipset's and Stein Rokkan's lead, privileging social cleavages as the main features influencing party system development. Recently, scholars have attempted to reconcile or integrate the competing approaches. Pippa Norris's new book fits into this trend by evaluating the two schools of thought side by side, harnessing valuable data produced by the Comparative Study of Electoral Systems (CSES).

Norris, a prolific researcher, covers a substantial amount of terrain in Electoral Engineering. She classifies electoral rules and addresses voting behavior and representation. She integrates these topics into a coherent volume by retaining her main theme: assessing the relative strengths of rational choice and cultural modernization theories in explaining how elections influence political competition. She ultimately concludes that institutional features consistently affect voting behavior and representation. But cultural factors play an important role in some circumstances.

In the first part of the book, Norris establishes her research questions, specifies the hypotheses, and presents a taxonomy of election rules. Chapter 1 succinctly outlines how rational choice institutionalism and cultural modernization explain the relationship between rules and outcomes, and sets forth a series of hypotheses that are featured in her later empirical tests. She also outlines the strengths and weaknesses of CSES data. In Chapters 2 and 3, the author turns to electoral systems. She divides the globe into four main groups: countries employing majoritarian, proportional, or mixed rules (“combined,” in her terminology), and those with no elections. In her classification, majoritarian systems include plurality and majority rules, alternative vote, bloc vote, cumulative vote, and the single nontransferable vote. Proportional rules encompass varied forms of party-list systems and the single transferable vote. Combined systems are subdivided into independent and dependent systems, depending on the linkage between allocation tiers. In the next chapter, Norris sets forth the criteria by which election systems may be judged, based on notions of adversarial and consensual democracy.

After establishing the definitions of election rules, the second section explicitly evaluates how rules influence voting behavior. Chapter 4 addresses the relationship between election rules and the size and shape of party systems. Norris finds overall support for Duverger's propositions; majoritarian rules tend to exert a reductive effect on the number of parties in competition. Chapter 5 turns to the relationship between rules and cleavages. She finds that election rules affect the strategies parties pursue; societies with majoritarian rules tend to exhibit weaker cleavage structures. Further, cleavages are stronger in the most modernized societies, contradicting expectations from the cultural modernization approach. This finding is repeated in Chapter 6. Party identification is weaker in majoritarian systems than in combined and proportional systems. Its effect is not significantly different between industrial and postindustrial polities, but postindustrial societies counterintuitively produce slightly more partisanship than industrial societies. In Chapter 7, Norris finds that election rules influence turnout, with voters coming to the polls more readily when proportional rules are employed. But factors important in cultural modernization theory, such as social features and cultural attitudes, also influence the likelihood of voting.

Norris evaluates how election rules influence representation in the third part of the book. In Chapter 8, she investigates women's access to political office. Proportional rules tend to benefit women, and gender quotas create conditions for seat acquisition. These institutional rules mitigate cultural features, represented by religion in the analysis. Ethnic minority representation is the focus of Chapter 9. The findings in this chapter are less conclusive, and the author speculates that the model's parsimony may contribute to her failure to find clear connections between institutional rules and outcomes. Chapter 10 concludes the analytical chapters by assessing accountability through modes of constituency service. As expected, majoritarian systems provide incentives for politicians to engage in personal, rather than party-based, appeals to voters.

The book is strongest when Norris employs data from the CSES, evaluating her research questions by using public opinion surveys from 32 countries. Indeed, CSES data provide new insights into variation in citizen attitudes across established and new democracies. But the book also suffers from some shortcomings. Combined systems are defined as one of the main families of electoral rules. Yet the hypotheses set forth clearly defined expectations only for majoritarian and proportional systems. The reader is left with no way to disentangle the effects of combined systems, except to assume that they are an intermediate form that should fall somewhere in the middle. The classification of electoral systems is based on a comprehensive survey published in 1997. Because election rules are a moving target, updating the classification to account for changes would have been a valuable contribution. Further, Belarus's 2000 presidential election is included in the analysis, although it is probably not equivalent to the other cases under investigation. The manuscript also features minor inaccuracies scattered throughout.

All in all, Electoral Engineering is a fine introduction to the main debates in the literature on comparative electoral systems and their effects. The book's strengths lie in its scope and ambition. Norris's exploration of CSES data provides scholars with seeds for future research projects. More advanced readers may leave wanting more, however. Norris warns readers of this possibility in the introduction, recounting how a colleague responded to her plans for a comprehensive book about elections. He simply noted: “It is complicated” (p. ix).